Skip to main content
Back to Library
Prompt Engineering Guide

Mastering Write email
on Phi-3.5 MoE

Stop guessing. See how professional prompt engineering transforms Phi-3.5 MoE's output for specific technical tasks.

The "Vibe" Prompt

"Write an email to a colleague, Sarah, about an upcoming project meeting. Make sure to include the date, time, and agenda points. Keep it clear and concise."
Low specificity, inconsistent output

Optimized Version

STABLE
As Phi-3.5 MoE, your task is to compose a professional email following a structured thought process. Aim for clarity, conciseness, and completeness. **Thought Process:** 1. **Identify Sender and Recipient:** Sender is [Your Name/You], Recipient is Sarah. 2. **Determine Purpose:** To inform Sarah about an upcoming project meeting. 3. **Extract Key Information to Include:** * Event: Project Meeting * Date: [Date of Meeting, e.g., 'Monday, October 26th'] * Time: [Time of Meeting, e.g., '10:00 AM PST'] * Location (if applicable): [Location, e.g., 'Conference Room 3B' or 'via Zoom link'] * Agenda Items: [List 2-4 concise agenda points, e.g., 'Review Q3 Performance', 'Discuss Q4 Strategy', 'Action Item Assignment']. 4. **Structure the Email:** * **Subject Line:** Clear, concise, and informative (e.g., 'Project Meeting: [Project Name] - [Date]'). * **Salutation:** Professional (e.g., 'Hi Sarah,', 'Dear Sarah,'). * **Opening:** State the purpose directly. * **Body:** Present the essential details (date, time, location, agenda) in an easy-to-read format (e.g., bullet points for agenda). * **Call to Action/Next Steps (Optional but good practice):** e.g., 'Please come prepared to discuss...' or 'Let me know if you have any questions.' * **Closing:** Professional (e.g., 'Best regards,', 'Thanks,'). * **Signature:** [Your Name]. 5. **Review and Refine:** Check for clarity, conciseness, grammar, and completeness. Ensure all requested information is present. Remove any superfluous words. **Constraint:** The email must be no more than 100 words. **Output:** Generate the email based on the above thought process and extracted information, using placeholder values where specific data is not provided, clearly indicating these placeholders (e.g., [DATE]).
Structured, task-focused, reduced hallucinations

Engineering Rationale

The optimized prompt provides a explicit, step-by-step chain of thought, guiding the model through the email writing process. It breaks down the task into manageable sub-tasks, ensuring all necessary components are included and structured correctly. By specifying roles, purpose, required information, and even a template structure, it reduces ambiguity and the cognitive load on the model, leading to more consistent and higher-quality outputs. The 'Constraint' explicitly sets a word limit, further enhancing conciseness. Explicitly prompting for placeholders ([DATE]) when specific data is missing helps the model maintain structure even with incomplete input.

0%
Token Efficiency Gain
The generated email from 'optimized_prompt' must logically follow the defined 'Thought Process'.
The generated email from 'optimized_prompt' must include placeholders for any missing information (e.g., [DATE], [TIME]).
The 'optimized_prompt' must explicitly mention 'Chain-of-Thought' or guide the model through a step-by-step process.

Ready to stop burning tokens?

Join 5,000+ developers using Prompt Optimizer to slash costs and boost LLM reliability.

Optimize My Prompts